Introduction
Recent attention to a new strategy of gender mainstreaming (GM) or, as
it is known in Canada, gender-based analysis (GBA), has taken hold both
internationally and nationally. This strategy, formally adopted at the
1995 Fourth World Conference on Women in Beijing, illuminates the
significance of gender as a central element of thinking and acting (Vlassov and Garcia Moreno, 2002) and recognizes its
role in power relations and institutions (Woodward,
2003). GM assumes that women and men are differentially affected by
policies and its aim is to integrate such knowledge into all dimensions of
decision making. GM requires that from inception all policies should be
analyzed for their gendered impact so that they can benefit men and women
equally. It has been hailed as a “potentially revolutionary
concept” (Pollack and Hafner-Burton,
2000), a significant policy innovation, and even a paradigm shift
for thinking about gender equality in policy-making processes (Rees, 2002). Despite the rhetoric about GM's
potential, its impact has been uneven. Not surprisingly, many feminist
scholars have been critical of the mainstreaming strategy and have been
reluctant to contribute academically to the policy debate regarding GM
(Booth and Bennett, 2002; Bennett, 2000). Consequently, an unreflective
interpretation of GM continues to be held by both policy makers and
femocrats (Carney, 2004). The discursive effects
of GM on constructions of gender and equality are not being interrogated.
In particular, the potential of recent feminist theory for providing
conceptual and analytical knowledge of the complex circumstances involving
gender differences and intersectionalities and multiple identities remains
largely uninvestigated.
In this paper, I provide an overview of GM, focusing on its
conceptualization, political context and the challenges typically
identified as impeding its effective implementation. I suggest that one of
the most overlooked impediments to GM's growth and impact is its
present disconnect with its feminist theoretical groundings. Contemporary
feminist developments in understanding gender and the interface between
gender, race, class, nationality, ethnicity, sexuality and power are not
adequately reflected in the concept of GM or in the strategies and tools
that have been developed to engender public policy. In its attempts to
institute social justice, GM has not moved beyond the male-female
dichotomy so prevalent in second-wave liberal feminist theorizing. As a
result, GM has become a watered-down approach to challenging the status
quo. This in turn affects how gender issues are constructed and leads to
important issues being excluded or marginalized in the realm of policy.
And yet, it is the very knowledge embedded within present-day feminist
theory, and in particular around equality/difference and diversity
debates, that is essential to being able to ask the right questions, to
develop the right approaches, and to anticipate intended and unintended
consequences of policy decisions.
The argument of the paper is that the relationship of feminist theory
and practice needs to be revisited if the mainstream is to be transformed.
In making this argument I am contributing to the literature theorizing
mainstreaming, which is still at a rudimentary stage of development (Booth and Bennett, 2002). By linking theory and
practice I am also responding to Iris Marion Young's call that
feminist theorists should take a more pragmatic orientation to
intellectual discourse by “categorizing, explaining, developing
accounts and arguments that are tied to specific practical and political
problems” (1994: 717–718).
Accordingly, my project is driven by the current problems linked to the
theoretical shortcomings of GM, which have practical importance in terms
of social justice in the realms of policy, research and practice. Narayan
and Harding (2000: vii) have gone as far as to claim that analyzing policy
in this way is crucial because “the shape of the conceptual
frameworks that guide public policy can be a matter of life and
death.”
If we take seriously the need to apply the insights of recent feminist
theorizing, it becomes clear that there are in fact no real possibilities
to adequately improve or expand the GM framework. GM is inherently limited
and limiting because it always prioritizes gender as the axis of
discrimination and moreover, the conceptualization of gender that GM rests
upon is clearly outdated. What is required is a broader approach to
mainstreaming, one that is able to consistently and systematically reflect
a deeper understanding of intersectionalities—the combination of
various oppressions that together produce something unique and distinct
from any one form of discrimination standing alone. Arguably, those who
are interested in developing effective mainstreaming strategies for public
policy can no longer be impervious to factors that are more important or
compound experiences of gender. However, as long as GM is used, this
primacy will not be displaced and the unique vulnerability of differently
socially constructed groups of women and men will remain obscured.
As a way to move beyond the current impasse in GM, I am proposing a
diversity1
In referring to the alternative as
diversity mainstreaming I want to acknowledge Bhabha's point that
diversity can depoliticize power differences and reduce difference to
simplistic versions (1994: 31–2). In the
current policy context, it is my contention that diversity is an expedient
term for policy discourse.
mainstreaming framework that draws on
Iris Marion Young's notion of “gender as seriality”
(
1994) and Kimberlé Williams
Crenshaw's work on intersectionalities (
2000,
1991). This framework
also builds upon a number of alternative models of mainstreaming developed
in the United Kingdom (
Donaghy, 2004;
Beveridge and Nott, 2002;
Rees,
1998). Diversity mainstreaming retains the category of gender,
albeit in a qualified manner. Most importantly, it puts front and centre
various forms of oppression (e.g., race, class, ethnicity, ability,
sexuality) and explores how they interconnect and mutually reinforce one
another. The framework, as I briefly demonstrate using the examples of
HIV/AIDS and human trafficking, is able to better capture, articulate
and make visible the relationship between various kinds of compounding
discrimination and oppression. This kind of analysis is currently absent
in GM but is of critical importance if decision makers are to create
effective policy. In sum, diversity mainstreaming provides what Dhamoon
has described elsewhere as a “roadmap for policy with normative
concerns for social justice” (
2004).
Gender Mainstreaming and its Political
Context
While it is important to acknowledge that there is no single
definition of GM (Mackay and Bilton, 2000; Woodward, 2001; Rees, 1998),
GM can be understood as “a deliberate and systematic approach to
integrating a gender perspective into analysis, procedures and
policies” (OECD, 2000). The goal, driven
by social justice (Rees, 2002), is to take into
account gender in all aspects of policy making by focusing on the adverse
effects of policy on both men and women and to address and rectify
persistent and emerging disparities between women and men (True and Mintrom, 2001). A useful definition, often
referred to in the literature, is that of the Economic and Social Council
of the United Nations:
Mainstreaming a gender perspective is the process of
assessing the implications for women and men of any planned action,
including legislation, policies or programmes, in any area and at all
levels. It is a strategy for making women's as well as men's
concerns and experiences an integral dimension of the design,
implementation, monitoring and evaluation of the policies and programmes
in all political, economic, and societal spheres so that women and men
benefit equally, and inequality is not perpetuated. The ultimate goal is
to achieve gender equality… .2
United
Nations ESCO (E/1997/L.30 Para Adopted by ECOSOC
14.7.97).
Mainstreaming gender often includes gender-sensitive as well as
women-specific policies and programmes. In fact this dual-track strategy
has been recommended as necessary for developing a comprehensive approach
to gender equality (Pollack and Hafner-Burton,
2000; Greaves and Hankivsky, 1999).
Moreover, mainstreaming strategies can be either
“integrationalist” or “agenda setting” (Jahan, 1995). The integrationalist approach seeks to
introduce a gender perspective into existing policy while an
agenda-setting approach seeks to challenge and transform policy paradigms
in the process of engendering policy. For the most part, agenda-setting
approaches are now being favoured over those that seek to address gender
issues within existing policy paradigms. There is an understanding that
the male-stream and androcentricity of policy (Beveridge and Nott, 2002; Rees,
1998) need to be challenged for meaningful transformation to be
realized.
Integral to the agenda-setting approach is the attention to not only
the specific interests of both men and women but also to issues concerning
different population groups, including the interrelated conditions and
factors that influence equality across the population (Health Canada, 2003). In general, GM typically
acknowledges that gender does not operate in isolation but in relation to
other factors such as race, ethnicity, ability, age, sexual orientation,
geographic location and so on. In fact, there is an assumption that the
ways in which GM advances gender equality are “equally accessible
and applicable to other areas of inequality” (Booth and Bennett, 2002: 431) and have “the
potential capacity to move beyond gender into other dimensions of equality
such as race and disability.” As the United Nations maintains,
“a strong continued commitment to gender mainstreaming is one of the
most effective means for the United Nations to support real changes at all
levels” (Hannan, 2001: 7).
GM models can be participative-democratic or expert-bureaucratic
(Nott, 2000). Participative-democratic models,
such as those developed in the UK, rely on civil society participation and
attempt to incorporate multiple equality areas (Donaghy
2004). Expert-bureaucratic models, like the approach embraced in
Canada, rely on gender experts being located in government bureaucracy. A
formal commitment to GM in Canada was made in 1995 when the government
announced its agenda in the document Setting the Stage for the Next
Century: The Federal Plan for Gender Equality (1995–2000),3
Although a promising development, it is
important to note that this policy shift, ironically, occurred at the same
time that national funding of women's groups was being reduced at
significant levels. It is after all, through the use of gender
mainstreaming that the diverse gendered damages caused by policies can be
brought to light and named (Teghtsoonian, 2000:
111).
which had been prepared for the Beijing conference by
Status of Women Canada, in collaboration with 24 federal departments and
agencies. In the
Plan as well as the more recent
Agenda for
Equality (2000), the federal government committed to ensuring that
all future legislation and policies include, where appropriate, an
analysis of the potential for differential impacts on men and women.
Challenges of Implementation
GM's promise to transform every policy process so that gender
equality is promoted in all planning and decision making has not been
realized in any jurisdiction or in any area of public policy (Hankivsky, 2005; Sjoroup,
2001; Woodward, 2001; Bretherton, 2001; Beveridge et al.,
2000; Burt and Hardman, 2001; Nott, 2000). In part this may be due to the fact that
GM is intended to be a long-term strategy. However, its failure to date to
bring about substantive change has led some observers to question its
role, as does Einarsdóttir (2003: 1):
“Why don't we see more progress in gender equality with all the
institutional, governmental, national, international gender equality
machinery we have to pursue our goals?” In the Canadian context,
limited progress on issues such as child care, unpaid work and pension
reform are cited as examples of the few inroads femocrats have had in
instigating gender-sensitive policy (Chappell,
2002: 100). Recently the Commonwealth Secretariat noted that
“despite a much greater level of overall awareness on gender issues,
and despite the solid body of research and analysis that now exists,
policy and programmes continue to show very limited and compartmentalized
concerns with gender equity” (Kabeer,
2003: 24). Indeed, the necessary conditions and components for
radical transformation in policy continue to be debated (Woodward, 2001).
Even though implementation remains highly variable across states,
analyses of GM to date tend to concentrate on best practices and,
alternatively, the political, legal and institutional barriers and
obstacles to effective implementation. The need for a supportive policy
environment has been recognized (Health Canada,
2003; Status of Women Canada, 2002; Squires and Wickham-Jones, 2002). For example, in the
postwar citizenship regime in Canada, the goals of social justice and
equity were accepted and therefore, claims made by groups such as women
were seen as part of the political mainstream (Jenson
and Phillips, 1996). Now one can argue that GM is at odds with the
neoliberal focus on individuals and suspicion of identity-based politics
(Teghtsoonian, 2000: 110). Neoliberal
priorities, including privatization and deregulation, are not conducive to
protecting or promoting women's equality. Nor do they provide the
convergence of ideas and interests necessary for the institutionalization
of GM. Bretherton (2001) refers to the lack of
such convergence as “swimming against the tide.” For GM to be
effective, its goal of gender equality must be culturally approved (Bustelo, 2003; Squires and
Wickham-Jones, 2002). It must also resonate with the values and
norms of regular actors in the policy process (Verloo,
2001). As Stone puts it, the policy agenda is determined largely by
a complex interplay of ideas and values that can be emotionally and
ideologically laden (Stone, 1989).
Second, governments at all levels need to prioritize gender issues if
the mainstream is to be successfully transformed (Bretherton, 2001). This includes the proper resourcing
of GM initiatives, as attempts to mainstream typically tend to be
under-resourced (Rees, 2002). Third, the
importance of having GM integrated in all steps of the policy process has
been highlighted. As Burt and Hardman emphasize (2001: 210), “[i]f GBA is applied in
the middle of the policy cycle, after a policy direction has been set, it
can have only a limited impact on the shape of policy.” Fourth, the
need for a diversity of methodologies and tools for the range of policy
sectors has also been recognized (Status of Women
Canada, 2002). Fifth, a more bottom-up approach to GM that includes
the insights of the women's movement has also been identified.
Specifically, consultation with a range of women's organizations and
interests has been determined as a key requirement of GM (Stetson and Mazur, 1995). Finally, the need for
effective state mechanisms and adequate training among government
bureaucrats is seen as essential to successful GM. At the same time, GM
should also be “user-friendly” (Booth and
Bennett, 2002). As Woodward has argued, GM “should be
something that can be learned and carried out by the Weberian ideal
typical androgynous servant of the state” (2001: 70).
Theoretical Issues
While addressing the practical requirements, including the political
and institutional contexts, is important, it is equally, if not more,
important to examine the theoretical framework that informs the
strategies, techniques and tools of GM. Examinations of such nature have
been largely overlooked in the literature and practice of GM. As Beveridge
et al. have argued elsewhere, “[t]here has been little
attempt to develop a general theory of mainstreaming which transcends the
diversity of state practice in order to provide a universal frame of
reference, or set of criteria, by which mainstreaming may be understood
and particular mainstreaming initiatives judged” (2000: 388). Similarly, Booth and Bennett have noted,
“the literature theorizing mainstreaming is still at a rudimentary
stage” (2002: 432).
This gap may be somewhat surprising, given the links between GM and
feminist theory. While it is often argued that GM emerged from networking
and promotion of the strategy by femocrats at the Beijing conference
(Russell and Sawer, 1999; Sawer, 1996), less attention has been paid to how
mainstreaming as a concept was transferred from the realm of feminist
theory to policy application (Carney, 2003;
True and Mintrom, 2001; Woodward, 2001). In particular, feminist theories
about engagement with the state and normative arguments regarding
women's oppression, subordination and inequality constitute the
foundation on which GM is constructed (Carney,
2004). By looking to theory, the challenges of engaging with the
state to bring about social change can be grasped, and the core content of
gender mainstreaming can be interrogated. We need to ensure, as Verloo
puts it, “a more dynamic connection to feminist academic
knowledge” (2001: 17).
Engagement with the state
Mainstreaming is about working within the system while at the same
time criticizing it (Einarsdóttir, 2003).
It is not unlike other struggles in which feminists have outlined both the
limitations and possibilities for social change when interacting directly
with the state. In the case of GM, this presents particular challenges.
While the gender equality machinery of the state “needs the
theoretical knowledge of feminism, the very substance of that same
knowledge has to be assimilated or ‘translated’ into the
language of the establishment, in order to be negotiable” (Einarsdóttir, 2003). Determining the proper
language to use in GM often entails reducing and even distorting gender
equality to technocratic language. When the essence of GM gets “lost
in translation,” this of course raises the larger issue of whether
or not feminists should engage with the state when seeking fundamental
changes. Wendy Brown for example, cautions against a myopic over-reliance
on the state (1995). bell hooks has similarly
pointed to the difficulties of abandoning the safety of the margins when
one engages with the mainstream (1996). Working
inside the state does put one at risk for losing the perspective of the
outsider (Spalter-Roth and Schreiber, 1995) and
it can interfere with the ability to maintain a certain distance from
political events. Indeed, engagement with the state necessitates an
imperative of compromise and the need to adjust radical demands to those
that are politically feasible (Prugl, 2004: 6).
In turn, this may lead to a measure of dependence and implicit agreement
to abide by state rules.
Arguably, the potential for compromise or even cooptation is a real
problem for many feminist theorists and one of the reasons why there is
resistance by academic feminists in regards to practical work on issues of
gender equality (Bacchi, 2001). It is important,
however, to remind ourselves that feminism is both an intellectual and a
political movement that seeks justice for all women (Haslanger and Tuana, 2003). And if we accept that
“the state, as an area of political practice, is not something
feminists can choose or refuse to enter” (Marshall, 2000: 94), then we must continue to explore
ways in which the strategy of mainstreaming may be improved, while being
“alert not to be swept away by the mainstream” (Verloo, 2001). And finally, it is also worth noting,
as does Vickers, that despite the challenges “many feminists see
state institutions as potential allies that women can mobilize to help
them achieve the changes they want and need” (1997: 14).
One largely uninvestigated way that improvement can be realized is if
the questions and critiques being raised within modern feminist theory are
better reflected in mainstreaming to challenge the dominant paradigms of
this form of policy discourse. At the same time, it is crucial to
understand the perimeters in which both policy and theory operate and the
kind of relationship and conceptual bridges that can and conversely cannot
exist between these two disparate realms. First, there should be explicit
recognition that the realm of policy is fundamentally different from the
realm of theory. What is possible in theory is often not possible when one
attempts to translate theory to practice. Some practitioners, for example,
question the applicability of the knowledge produced by feminist theorists
(Einarsdóttir, 2003). Woodward (2001) has argued that adapting a “gender
approach” in GM helps to win broader audiences for gender issues
precisely because it is not associated with feminism or feminist theory.
Others have suggested that feminist theories are “ill-suited to deal
with real-world issues of human difference and diversity”(Armstrong, 2003).
In addition, it may not be possible to draw on theory in a linear
fashion to sketch out a coherent blueprint or pathway for realizing gender
equality. To some it may even seem paradoxical to look to feminist theory,
which is inconclusive in terms of its treatment of gender and equality, to
improve GM, which seeks clarity in promoting gender equality. Despite this
tension, however, insights from theory do provide the impetus for
rethinking the efficacy of the current framework of GM. While there is
much to draw upon from theory to advance mainstreaming, the reverse is
also true. There are shortcomings in terms of how feminist theorizing
approaches questions of gender and equality. In many ways, this becomes
more evident when one attempts to translate theory to practical
application. Often, however, feminist theorists who interrogate issues of
difference and diversity do not engage in practical debates or political
activism (Squires, 1999: 136). So while the
project of moving the debate on gender and social justice forward may be
compelling in theory, if we are to follow theory to what I think should be
one of its logical conclusions, that of being able to deal with
real-world issues, we may find that certain adjustments may need to be
made to the theory itself. Here I am in full agreement with Carens'
(2000) argument that we really do not understand
theory until we see it interpreted and applied in a variety of
contexts.
Current GM Framework and its Normative
Underpinnings
Before I move on to demonstrate what transformation may come about by
connecting GM to recent feminist theoretical developments, it is important
to look at the status quo and current practices. Rather than continuing to
evolve in a parallel fashion with the developments in feminist theory, and
particularly those that have emerged in response to the second wave of the
women's movement, GM, in fact, has been frozen in its content and
state of knowledge (Verloo, 2001). It has
‘taken on a life of its own.’ GM's potential to bring
about the normative transformation of the existing social and political
order that leads to gendered outcomes (True and
Mintrom, 2001) has been hampered by its growing disconnect from the
valuable insights of contemporary feminist theory. It has in essence
become a strategy, a set of methods, without an adequate theory. I
therefore disagree with those who claim that GM corresponds to the most
recent developments in academic feminism, and in particular to the
significance of diversity and intersectional types of oppression (Einarsdóttir, 2003). In fact, questions and
critiques being raised within well-elaborated feminist theory are not
systematically or consistently reflected in GM.
The concept of mainstreaming has become synonymous with the gender
perspective alone (Booth and Bennett, 2002) and
within the context of policy decision making, this perspective promotes a
very one-dimensional conceptual understanding of gender equality. Burt and
Hardman (2001) are correct in observing that
gender and equality are highly contested concepts whose complexity is not
always reflected in GM. As gender has become a more conventionally
accepted term and concept, GM has divorced itself from the “feminist
transformatory project” (Baden and Goetz,
1997: 7). Disjointed from the progress made in feminist theory, GM
invokes a liberal concept of an abstract woman, obscuring the variety of
conditions that inform women's experiences, needs and status. As
Verloo (2001) elaborates, the
phrases—inequality between men and women, differences between men
and men, equal opportunities between men and women—are used without
a clear understanding of what they entail and how they relate to the goal
of gender equality.
Significantly, despite its intention to attend to diversity, GM tends
to concentrate on differences between men and women, treating each gender
as a unitary, one-dimensional category of analysis, further obscuring the
differences among and between women (Beveridge et al.,
2000; Burt and Hardman, 2001) and in
particular, the variety of conditions that inform women's
experiences, needs and status. The focus on differences between men and
women also obscures considerations and analyses of power and inequality
(Marshall, 2000). This is most clearly evidenced
in the manuals, measurement techniques and tools used to measure gender
equity. Gender-disaggregated data, gender equality indicators, gender
impact assessments, gender proofing and engendering budgets are focused
predominantly on fairly crude distinctions between women and men. In many
ways, this simplistic way of delineating between the sexes demonstrates
the ability of GM to cohabit with liberal political and economic
structures, and its inability to provide the radical critique of existing
power relations (Carney, 2004: 19) necessary for
social justice.
The central problem is that GM prioritizes a coherent analytic
category of gender, in which race and class, among other factors, are
considered as an add-on to gender. For example, in Canada, where
this approach is particularly evident in the manual put out by Status of
Women Canada (1996), we are given weak
analytical tools for understanding women's diversity, such as:
consider how experiences of women and men will differ geographically, and
are influenced by poverty, colour, aboriginal ancestry,
disability/ability. Health Canada's Gender-Based
Analysis states that the gender-based framework should be overlaid
with a diversity analysis. Similarly, the policy of Human Resources
Development Canada (now divided into Social Development Canada and Human
Resources and Skills Development Canada) calls for policy analysts to
“consider gender together with other demographic/diversity
factors” (2003: 13). The Canadian
International Development Agency's Policy on Gender Equality
contends that the use of gender analysis provides information on
“the difference among women and men and the diversity of their
circumstances, social relationships, and consequent states” (for
example, their class, race, caste, ethnicity, age, culture and abilities)
(CIDA, 1999:17), without adequately
incorporating these considerations into its implementation strategies. And
although the Gender Equality Analysis of Indian and Northern
Affairs emphasizes the importance of considering diversity, diversity is
listed as a factor “in addition to gender” (1999: 6). Teghtsoonian has similarly argued that
“despite drawing attention to the specific circumstances of
multiply-marginalized women, the focus in these documents tends to remain
on gender-in-general” (1999: 5).
Even if the intention of GM is to have the needs of different
demographics and specific groups considered (Donaghy
2004), this does not occur as a matter of course with any
consistency. And, the often-resulting “add other differences and
stir” approach is ineffective. It is not appropriate for other
equality dimensions (Rees, 2002) or for
understanding multiple and diverse forms of oppression. When complex
relationships are simplified and systematically reduced to only one form
of discrimination, as in the case of gender discrimination, there is a
serious risk that the causes and consequences of these experiences will
remain misconceived and inadequately addressed. Harding notes that this
kind of approach faces the “same kinds of problems encountered by
attempts to add women and gender to conceptual frameworks designed,
intentionally or not, to exclude them” (1998: 32). Thus, the current GM framework reveals the
gap between feminist practice and feminist theory. Moreover, policy
interventions that are based on such incomplete knowledge of women's
lives cannot be effective.
The obvious problem, of course, is that women are not a single
constituency with the same social and cultural backgrounds. Not all women
who live within the same society at any given point in time are oppressed
or subjugated in the same way (Yuval-Davis,
1997). Gender is interlocked with class, race, ethnicity and other
structural relations that underpin a society's institutions and
practices (Harding, 1995). Thus, tolerance of
differences and attempts to include diversity do not go far enough in
changing GM. In reality, in many instances, factors other than gender are
the primary cause of discrimination, oppression and inequality. The
project of transforming the mainstream therefore “runs headlong into
the question of what women's needs are and which
women's needs they are” (Kittay,
1998: 573). Important questions that GM has not adequately grappled
with include: What is gender? What is equality? Can we speak of women as a
group? What constitutes salient issues for GM and who defines such issues?
How can differences among women be dealt with in formulating, implementing
and evaluating policies? Returning to the roots of GM, to feminist
theorizing, can inform the process of beginning to address some of these
key issues.
Feminist Theory—Gender and
Equality
Recent debates about gender, and gender equality, within feminist
theory have been complicated. Difference feminism (MacKinnon, 1987; Hartsock,
1987; Irigaray, 1993; Kristeva, 1984), diversity feminism (Lorde, 1984; Collins, 1991;
Mohanty, 1997; Haraway,
1990; Anzaldúa and Keating, 2002)
and deconstruction feminism (Butler, 1990; Fraser and Nicholson, 1990) represent various currents
in the theoretical debates over questions of gender, identity,
intersectionality and equality. What is clear is that feminists are
grappling with the complexities of gender questions, the problematic
nature of gender construction, and the need to address women in their
diversity. Diversity feminism, in particular, seeks to recognize and
empower traditionally marginalized, oppressed and excluded women. One of
the primary debates within feminist theorizing during the last decade has
been whether it is possible or desirable to posit a category such as
women, upon which feminism is predicated and to which it responds. Without
doubt there is growing skepticism about the value or relevance of
gender-specific theoretical outlooks. Included in the theoretical
skepticism is the very concept of gender and its relation to the category
of women. Mary Dietz (2003) puts it effectively
when she argues that feminist theory is struggling to identify “the
critical conceptual coordinates” of its claims. And, the stakes in
current feminist theoretical debates are “every bit as practical and
political as they are categorical, analytical, and philosophical”
(403).
As a result of these recent theoretical developments, we must
seriously consider whether a focus on gender is adequate for understanding
inequality or for moving towards social justice in public policy. As
Kittay explains elsewhere, “[t]he seeds of transformative
possibilities in social policy affecting policy may in fact be found by
looking beyond gender as such” (1998: 54).
Without any doubt, we need a more critical understanding of gender and
gender equality, beyond what is currently reflected in GM literature and
practice. Most importantly, it is clear from feminist theory that simply
recognizing or attending to diversity and difference is inadequate for
understanding the texture and diversity of women's lives. The
confluence of factors that compound one another to produce discrimination
must be integrated as a fundamental basis of feminist theory and, by
extension, of GM. But as Verloo has observed correctly, “the
relationship between gender and ethnicity/or race, between gender and
sexuality, or between gender and class, to name just three of the most
important structural inequalities are too complex for the current
conceptual framework of gender mainstreaming” (2001: 21).
Even though improvements to GM have been proposed (Booth and Bennett, 2001), some feminists have
questioned whether or not factors that create gender equality are distinct
enough to warrant a specific GM approach or whether a broader approach to
mainstreaming equality is required (Verloo,
2000). As a result, mainstreaming equality has been both proposed
and applied in a number of jurisdictions (Donaghy,
2004; Beveridge et al., 2000; Mackay and Bilton, 2000). As Donaghy explains, in
Northern Ireland and Scotland, “the concept of mainstreaming is
shared by GM and mainstreaming equality, and the difference between the
two models is the groups for which equality of opportunity is promoted.
Gender mainstreaming considers men and women, while mainstreaming equality
adopts a number of groups for whom equality of opportunity is
considered” (2004: 51). Support for a
race-based parallel to gender mainstreaming has also been expressed (Williams Crenshaw, 2000). But as Mackay and Bilton
correctly observe, “further theoretical work is also needed …
to construct more sophisticated understanding of
‘simultaneous’ experience of different dimensions of
difference and disadvantage” (2003:
153).
To move the analysis forward in any meaningful way, it is useful to
begin where theory and practice converge. The common challenge, which is
particularly salient in terms of GM, is well articulated by Arneil when
she asks: “[Is] it possible, or desirable, to maintain a
unified force of ‘women’ and their different perspective(s) in
order to resist the historical oppression of women, as women, while
simultaneously incorporating, in a serious way, the
‘differences’ among women in both theory and practice?”
(1999: 204). From the valuable insights gleaned
from theory, seriously incorporating the differences among women in
practice requires more than naming, being aware or considerate of these
differences and their relationship with one another. This is where GM
falls short and is inherently inadequate. The key issue, as Teghtsoonian
puts it, is “how initiatives to ‘mainstream gender
analysis’ can respond to diversity among women in a positive and
productive way” (1999: 5).
What is needed at this juncture in time is a conceptual shift that
will lead to a strategy that is far more progressive and consistently
inclusive. What I propose is a diversity mainstreaming framework that
transcends the current focus and form of GM. While other jurisdictions,
including Northern Ireland, Wales and Scotland, have opted for models that
resemble such an approach, diversity mainstreaming represents a new
direction in the Canadian context. In elucidating the diversity
mainstreaming approach, I put forward a qualified defence of gender, but
one that challenges traditional conceptualizations and displaces gender as
the primary axis for understanding experiences of discrimination,
inequality and oppression. Accordingly, the privileging of gender is
replaced with a more sophisticated and comprehensive approach to
understanding lived experiences, especially for those persons who have
been and continue to be marginalized by current perspectives and practices
of GM. Most importantly, the diversity framework places the importance of
intersectionalities front and centre in the analysis. This is consistent
with Collins' insistence that “all must support a working
hypothesis of equivalency between oppressions that allows us to explore
the interconnects among the systems and extract us from the internecine
battles of whose oppression is more fundamental” (2002: 83).
Diversity Mainstreaming
Qualified Defense of Gender (and Sex)
In formulating the diversity framework I do not dismiss categorically,
as do some feminist theorists, the category of gender. Also, unlike many
feminist theorists, I do think that the category of sex is also
important—in particular in the realm of health policy.4
Sex can determine differential propensities for
certain health conditions or diseases, different risk factors and
treatment requirements (Greaves, Hankivsky et al.,
1999).
Drawing on Young's work on gender as serial
collectivity, I am putting forward a
reconceptualized concept of
gender that challenges the essentialization associated with gender. So
while I am not altogether moving beyond gender as such, I am proposing a
more nuanced and sophisticated understanding and I am seeking to
appropriately situate gender within diversity politics and public policy.
My position in relation to gender is similar to that of Di Stefano, who
argues that “gender is basic in ways that [we] have yet to
fully understand, that it functions as a ‘difference that makes a
difference’ even as it can no longer claim the legitimating mantels
of
the difference” (
1990:
78).
Within this alternative conceptualization, gender remains an important
category of analysis.
I agree with Young that “there are pragmatic political reasons
for insisting on the possibility of thinking about women as some kind of
group” (1994: 714). Gender has practical
implications for both women and men, and thus in the context of policy
remains important interpretively and politically. It is clear that
“where gender has not been insisted upon as a category of analysis,
gender-blindness is the result” (Marshall,
2000: 67). Burt's analysis (1995) of
the articles published in Canadian Public Policy between 1975 and
1993, in which only 1 per cent of published works focused on women's
issues, while mainstream policy analyses were void of any kind of gender
analysis, illustrates the dangers of such gender blindness and exclusion
of women from policy agendas and processes. The question remains, however,
how to bring to the foreground issues without essentializing gender.
As feminist theorists have made very clear, we need to find a way to
resist and challenge the normalization and essentialization that occurs
when we try to describe women as a group, experiencing “sameness of
oppression.” In her response to the challenges of difference,
Phillips has argued that “in the reworking of contemporary political
theory and ideals, feminism cannot afford to situate itself for
difference and against universality, for the impulse that takes
us beyond our immediate and specific difference is a vital necessity in
any radical transformation” (1993: 71).
Phillips' position is not altogether satisfactory because it does not
lead to a critical examination of the conceptualizations and implications
of universality constructs. It is extremely problematic to assume, as
Mohanty has noted, that women are “an already constituted, coherent
group with identical interests and desires, regardless of class, ethnic or
racial or locations, or contradictions” (1997: 80).
Young has suggested one way out of this dilemma, which has particular
significance for my approach to reconceptualizing mainstreaming, in her
examination of gender as seriality. Drawing on the concept of class
seriality that Sartre develops in his Critique of Dialectical
Reasons, Young proposes “a way of thinking about women as a
social collective without requiring that all women have common attributes
or a common situation” (1994: 723). Her
approach points to the need for “some conception of women as a group
prior to the formation of self-conscious feminist politics” (Young, 1994: 722). According to Young, serial
collectivity provides a way of thinking about women without having to rely
on identity or self-identity. In a serial collectivity, Young maintains
that members are unified passively by the objects around which
their actions are oriented, limited and constrained, or by the objectified
results of the material effects of the actions of others.
Unity of the collectivity is shaped by what Young refers to as
practico-inert realities that construct gender. These are the material and
social facts that each individual must relate to and deal with (1994: 731). The practico-inert realities that
construct gender include female bodies, other objects and materialized
historic products which condition women's lives as gendered (Young, 1994: 729). These can be verbal, visual
representations, artifacts and social spaces, clothes, furniture,
cosmetics and so on. For Young, the material organization of social
relations as enabled and constrained by the structural relations of
enforced heterosexuality and the sexual division of labour position women
in a gender seriality. At the same time, she cautions that this represents
a “passive unity” and that in the end, she is only claiming
“that the level of gender as series is a background to rather than
constitutive of personal and group identity” (1994: 731).
Young is right when she concludes that “applying the concept of
seriality to gender makes theoretical sense out of saying that women is a
reasonable social category” (1994: 728)
without falling into the trap of essentialization and undifferentiated
analysis. Marshall similarly argues that “we do not need to believe
in any common essence to a category for it to have intelligible meaning,
even in a sense which recognizes that it is mutually constitutive with
other categories, as is always the case when we talk about gender”
(2000: 54). In Young's approach, gender
structures are not defining attributes of individuals because individuals
can relate to them in different ways in different social contexts.
Young's work, like that of a number of other feminists, seeks to
bridge the divide between essentialist feminism and those who see the
utility in maintaining some category of “woman.” With her
specific interpretation, Young does leave open the door to variability and
diversity in experience even if there are pratico-inert realities in every
society. In the end, however, Young's approach, although effective in
challenging the homogenizing tendencies of gender constructs, is
incomplete since it does not necessarily lead to any kind of
meaningful intersectional analysis, where gender does not always prevail
“over and above everything else” (Mohanty,
1997). I am therefore in complete agreement with Williams Crenshaw
when she argues that “while it is true that all women are in some
way subject to the burdens of gender discrimination, it is also true that
other factors relating to women's social identities such as class,
caste, race, colour, ethnicity, religion, national origin and sexual
orientation are ‘differences that make a difference’”
(2000: 3). The fact that universal gender
inequalities exist does not mean that gender oppression is more oppressive
than racism, classism, heterosexism, ethnocentrism or exploitative global
economic conditions (Whittle and Inhorn, 2001:
160).
Moving Beyond Privileging Gender
The key is not to abandon the analytical category of gender but to
explicitly recognize that it is not the primary or key axis of social
oppression, or “the issue in GM” (Beveridge and Nott, 2002), and work towards a
mainstreaming framework that does more than add or attend to difference
while transforming the very policies and practices from which difference
emerges. Central to such an exercise is the “openness to
relinquishing genders' hegemony as a starting point for
analysis, looking instead to if and how it emerges as a significance in
particular circumstances” (Marshall, 2000:
162). Recently, Teghstoonian noted that “further work needs to be
done to weave systematically into gender analysis a focus on race, sexual
orientation and ability, and to weave gender into lens-based work focused
on other marginalized groups” (1999:
4).
What I am proposing, however, is different. It is not about fitting
missing pieces into existing frameworks. It is also different from the
approach in Northern Ireland,5
Section 75 (1)
of the Northern Ireland Act 1998
which is lauded as
“unique and world leading” because it expanded traditional
mainstreaming beyond gender (
Donaghy, 2004a;
McCrudden, 2003). The approach in Northern Ireland promotes a multiple
equality approach that requires due regard for the need to promote
equality of opportunity: a) between persons of different religious belief,
political opinion, racial group, age, marital status or sexual
orientation; b) between men and women generally; c) between persons with a
disability and persons without; and d) between persons with dependants and
persons without. At the same, it is less clear that this approach leads to
an analysis that effectively captures the relationships between all these
different and multiple equity grounds. The framework for diversity
mainstreaming is one that enables a truly integrated analysis, one that
systematically captures the interstices of all factors of oppression.
Diversity mainstreaming, thus conceptualized, builds upon Williams
Crenshaw's model of intersectionality. Crenshaw's approach
reflects the emerging literature, both normative and empirical, that is
seeking to deal with multiple grounds of disadvantage (Collins, 1991; Jhappan,
1996; Mouffee, 1993, 2000). This form of analysis “addresses
the manner in which racism, patriarchy, class oppression and other
discriminatory systems create inequalities that structure the relative
positions of women, races, ethnicities, classes, and the like”
(Centre for Women's Global Leadership,
2004). In so doing, it neither prioritizes nor essentializes gender
or race or any other factor. As Zierler and Krieger explain elsewhere,
this type of approach goes beyond “adding one-dimensional terms like
race/ethnicity or social class to a long list of
other variables … and looking for multiplicative effects. It
instead requires asking questions about deprivation, privilege,
discrimination, and aspirations, to permit characterizing people more
fully, and as more than the sum or product of their parts” (1995: 253). Similarly, Spellman has argued that
“one's gender identity is not related to one's racial and
class identity as the parts of pop-bead necklaces are related, separable
and insertable in other ‘strands’ with different racial and
class ‘parts’” (1988: 15).
Intersectionality recognizes this and instead seeks to illuminate the
synergistic effect of interlocking forms of oppression.
Williams Crenshaw uses the metaphor of a traffic intersection to
describe intersectional subordination. Race, class, gender and other forms
of discrimination, such as sexual orientation and ability, are the roads
that structure social, economic and political terrain. This metaphor
captures the numerous systems of subordination that often overlap and
cross and that create complex intersections. It allows us to avoid
thinking of these dynamics as disjoined or simply parallel. It is within
the intersections of these contexts that multiply burdened populations are
located. They must negotiate the oncoming traffic and the injuries from
the collisions of the various forms of inter-related and interlocking
discrimination, and decide whether these are pre-existing conditions or
brought on by their particular acts and policies.
Within a GM framework, where gender is dominant, these dynamics and
their consequences may be marginalized or completely invisible. In the
context of diversity mainstreaming, however, the mapping of multiple forms
of discrimination allows for the understanding of gender relations in
their specific context, and in particular, their relationship to other
structures of inequality such as class, ethnicity, nationality and sexual
orientation, among others. Using this approach can contribute to
furthering the variability of discrimination and oppression nationally and
internationally. Even though there is a cross-national convergence around
GM through the role played by transnational networks (True and Mintrom, 2001), strategies of engendering
policy at the national levels tend to obscure growing feminist critiques
regarding economic globalization (Hankivsky and Morrow,
2004). The complex and contradictory forces that operate on
national and global levels to determine and structure women's lives
must also be carefully investigated. If heterogeneity is to be respected
fully, women in the North, for example, should not be ignoring the
experiences and living conditions of women in the South. What would be
required, according to a diversity mainstreaming approach, is a
foregrounding of the intersections of the local and global (Hegde, 1998).
In the end, diversity mainstreaming allows for a more complex and
dynamic understanding of equality and social justice, because the contours
and compound effects of discrimination that women experience can be
captured and the invisibility or marginalization of differences is no
longer an option. Concretely, in terms of practice, diversity
mainstreaming would influence the following: the policy questions that are
asked, research design, development of improved research methods to
uncover key aspects of intersectional subordination, and data collection
(e.g., what kind of data is collected, how it is collected, and how it is
disaggregated). Using this framework would entail an explicit commitment
to take into account a range of women's perspectives. In the realm of
policy, this can be understood as bringing the voices of women, together
with their relevant concerns, to the formation, monitoring and evaluation
of public policy. This helps to avoid what Phillips has described as
policies that are worked out for rather than with
politically excluded constituencies (Phillips,
1995). Finally, diversity mainstreaming would change the way in
which policy analysis is undertaken and resultant policy is developed,
implemented and evaluated.
Practical Applications of the Alternative
Approach
HIV/AIDS
The need for diversity mainstreaming is apparent when considering the
multiple forms of discrimination and subordination that influence a
phenomenon like HIV/AIDS. It has been argued that GM is the most
effective and equitable way of using existing resources for combating
HIV/AIDS (Commonwealth Secretariat and Maritime
Centre of Excellence for Women's Health, 2002). To really
understand, contain and respond to HIV/AIDS, however, it is no longer
enough to frame our analyses solely in terms of gender. An analysis
prioritizing or focusing on gender would not necessarily capture
intersectional subordination and oppression, which are crucial to
understanding this pandemic. In comparison, diversity mainstreaming would
lead us to consider, for instance, patterns and prevalence of HIV
transmission that are region and country-specific; complex issues of
access to health services and medication including antiretrovirals,
differing rates of violence, cultural attitudes and norms towards sexual
behaviours and sexual orientation; and individuals' proximity to
centres of political, economic and social power. This analysis would not
subsume all experiences of oppression along the axis of gender, even
though gender could remain as one of a number of intersecting forms of
oppression. Instead it would consider all relevant factors as equally
important. It would reflect the reality that all aspects of HIV/AIDS
are embedded within the context of gender, race/ethnicity and class
oppression (Amaro and Raj, 2000). It would also
demonstrate how such factors cause oppression for those who are at their
intersection.
Trafficking
Another example through which the limitations of GM can be observed is
with the issue of trafficking. Trafficking is often described as one of
the most serious contemporary forms of gender discrimination. GM both
supports and perpetuates this framing of trafficking with its primary
focus on the analytical category of gender. As a result, reports by Status
of Women Canada and the United Nations pay little or no attention to the
role of race and other forms of subordination when assessing the risk for
being trafficked (Williams Crenshaw, 2000).
Interpreting trafficking of women as only an issue of gender
discrimination, while ignoring or not properly capturing the racial,
ethnic and class dimensions of the problem, harms women. This approach,
typical of GM, fails to properly take into account fundamental elements in
a proper analysis of causes and undermines the means for addressing the
problem (Hannan, 2001: 5). As Hannan
persuasively argues, “attention has to be paid to the groups of
women who are more likely to be trafficked and a greater understanding
developed on the links between their vulnerability to trafficking and
other aspects of their situation, such as race, ethnicity, religious
affiliation, etc.” (2001: 5). A fully
integrated analysis of trafficking would require that all factors that
contribute to the vulnerability of women in this context be included both
in the analysis of the problem as well as in the recommendations designed
to address the issues (Williams Crenshaw, 2000:
4). Diversity mainstreaming would be a way to ensure such an approach and,
moreover, would encourage weighing appropriately/equally all relevant
factors.
In the final analysis, any transformative potential of diversity
mainstreaming will still need to confront the constant barriers and
obstacles associated with state engagement and related political
struggles. Indeed, as Williams Crenshaw correctly notes, “efforts to
bring greater understanding of problems related to intersectionality face
a steep climb from its current invisibility to the conscious awareness of
… policy makers” (2000: 11)
Moreover, as Jhappan (1996) has noted, our best
efforts will never be able to include all relevant factors. These
challenges, however, should not undermine efforts to displace GM with
diversity mainstreaming. Diversity mainstreaming would allow for broader
inquiries about a whole host of issues extending beyond HIV/AIDS and
trafficking. By bringing to the foreground the various background
dimensions that interact to create layers of inequality and which
structure the relative positions of women and men, a more complete and
sophisticated analysis can be developed, one that better captures the ways
in which public policy is experienced by various groups of women and men
who may experience multiple discrimination. Policy makers may be persuaded
to incorporate this approach into their work if they understand that it
has the potential to lead to more effective, responsive and therefore
efficient policy decisions.
Conclusion
Without doubt, gender mainstreaming has brought some important
advances in terms of policy decision making, because it introduced the
idea that gender matters and that its differential effect must be analyzed
(Williams Crenshaw, 2000: 2). However, policy
development in this area is at an impasse. Despite good intentions by
governments and femocrats, GM is not transforming the mainstream. One key
area that is consistently overlooked in analyzing GM is its conceptual
framework. As illustrated in this paper, the GM framework is built upon an
outdated theoretical foundation that is able to capture at best partial
and distorted understandings of women's conditions in ways that tend
to be ineffective, inefficient and counterproductive (Williams Crenshaw, 2000: 6). If the promise of
GM—that is, social justice—is to be realized, there is a
pressing need to integrate present feminist knowledge about the context of
lives and experiences, structural inequalities in the public and private
spheres, and gender with all its intersectionalities.
The argument that I have made is that by returning to its theoretical
roots, it becomes clear that there is a disjuncture between GM and
contemporary feminist theory. GM is outdated and ineffective and needs to
be rejected in favour of a more expansive diversity mainstreaming
framework. The diversity mainstreaming framework proposed in the paper is
connected to feminist academic knowledge because it is able to take into
account emerging theoretical developments that provide crucial insights
into existing, complex and varying forms of inequities. It responds to
Verloo's urgent call “to formally develop more knowledge and
practices on the intersection of gender and other inequalities”
(2001: 22). In outlining the elements of the
proposed alternative, I have sought to illustrate a mainstreaming approach
that has the potential to broaden the terms of inquiry, lead to more
inclusive and egalitarian practices, affect the centre in profound ways
and, in the final analysis, contribute an important conceptual advancement
in expanding policy discourse in relation to social justice.
